Breathing Gas: Gas Purity Standards

Standards specify the composition and the maximum allowed concentration of contaminants in breathing air and for gases used in deep diving. Greater purity is demanded for gases used in deep diving because of the effect of higher inspired partial pressures for all gases in a mixture. Standards vary among organizations with respect to the acceptable level of contaminants, how these are to be detected and for which contaminants the gas should be tested.

Table 19.1 shows some of the available standards. Readers who may rely on these data should review the source documents for verification and more specific information. The standards are updated periodically, so it is important for users to monitor any changes and act accordingly.


Detailed consideration of the standards could lead to the opinion that most limits are rather conservative. Safety margins are incorporated for two reasons. First, the standards are generally based on extrapolation of the effects of the contaminants in isolation at 1 ATA (i.e. the surface equivalent value [SEV]). This may not be entirely valid for contaminants in combination at high pressures. Second, a safety margin will help to allow for any deterioration in the air quality among tests.

Standards are only lists of the maximum allowed concentration of some common impurities. Air may meet these specifications and still contain toxic substances. A greater variety of contamination problems occurs in caisson work where industrial equipment is being operated. Mineral dust from excavation and blasting is a common problem. Unusual contamination has also occurred in recompression chamber operations, especially if therapeutic or research equipment is used in the chamber. Because exposure times are generally greater in chambers, the toxic contamination has more time to exert its effect, particularly during saturation dives. Toxic substances that may be present include mercury from manometers, ammonia or Freon from leaking air conditioning plants, anaesthetic residues and other vapours from pharmaceutical preparations used during hyperbaric treatments. Potential hazards can be minimized by conducting an appropriate risk assessment.

In using gas mixtures prepared for deep diving, problems can result from the great pressure at which the mixtures are used. This will increase the risk of toxicity because of the higher partial pressures of contaminants. For example, and ignoring the dilution by water vapour in the lungs, at 1 ATA a carbon dioxide concentration of 2 per cent in inspired air means an inspired partial pressure (FICO2) of 15.2 mm Hg, and this is well-tolerated. At 5 ATA (40 msw), however, the same concentration means an FICO2 of 60.8 mm Hg – which will cause dyspnoea, increased work of breathing, distress and ultimately unconsciousness. The same problems will result from a concentration of only 0.2 per cent at 50 ATA. Therefore, standards need adjustment if they are used for depths and times greater than those assumed by their designers.

In the hyperbaric chamber environment (and elsewhere), toxicological problems can be introduced by the use of cleaning agents, among other sources. This is a growing problem with the rise in the prevalence of multi-resistant organisms that require extensive decontamination of therapeutic chambers between uses. The basic rule must be ‘if in doubt, leave it out’. Useful guidelines are available from reference to experience with long-term exposures in spacecraft and nuclear submarines. However, with the increase in hyperbaric treatment centres and greater reporting and communication among these centres, information on potentially problematic substances in a hyperbaric environment is more readily available.

The reasons for listing the components and the concentrations commonly specified are outlined in the following subsections:

Nitrogen and oxygen

The concentration of oxygen in compressed air standards is close to the level in clean, dry air. Any significant deviation is most unusual. If the nitrogen concentration was elevated, it could increase the risk of decompression sickness, narcosis or hypoxia. If the oxygen concentration was increased, the risk of oxygen toxicity, and fire hazards in hyperbaric chambers, would rise. The oxygen may be elevated by connecting to a bulk oxygen supply. This may be accidental, but it has been deliberate in the misguided belief that increasing oxygen concentration will increase the endurance available from a cylinder.

Carbon dioxide

A typical specified carbon dioxide level of 0.05 per cent (500 parts per million [ppm]) means that at 10 ATA the partial pressure of carbon dioxide would still be well below that required to cause any physiological effect. It has been argued that the British and Australian/New Zealand standards may be too strict because some other standards set a maximum limit of 0.1 per cent. This would not be toxic to the depth limits of compressed air diving and would be easier for compressor operators to meet.

The carbon dioxide level, even if it is within the specification used, should be considered in relation to the level in the ambient air. Global increases in carbon dioxide levels, further increased in cities and industrial areas, can cause compressor intake air to have carbon dioxide levels in excess of standards, even though they are physiologically safe. Some compressors use intake scrubbing of carbon dioxide from ambient air with absorbent canisters.

Carbon monoxide

This toxic gas binds tightly to the oxygen binding sites of haemoglobin to form carboxyhaemoglobin – preventing the carriage of oxygen. If sufficient haemoglobin binds with carbon monoxide, the diver will become hypoxic. The formation of carboxyhaemoglobin will also interfere with the transport of carbon dioxide away from the tissues by preventing its combination with haemoglobin. Carbon monoxide also causes oxidative stress and direct cellular toxicity.

The sometimes described cherry red colour of these victims is an unreliable clinical sign, especially in patients with cardiorespiratory impairment. Exertion and increased ventilation will hasten the development of symptoms. Subjects with a low haemoglobin level are more susceptible to carbon monoxide poisoning.

The concentrations required for poisoning are considerably greater than the maximum carbon monoxide level of 3 to 10 ppm specified in most standards. Exposure to ambient air with carbon monoxide levels higher that 100 ppm is considered dangerous to human health. A limit of 25 to 50 ppm is a suggested maximum level for occupational workers exposed for up to 8 hours a day.

For divers breathing air, the higher partial pressure of oxygen tends to protect against the effects of increased carbon monoxide partial pressure while at depth. The toxic limits of carbon monoxide at depth and how they are modified by varying ambient and oxygen partial pressures have not been established. Divers are probably at greatest risk of unconsciousness as they surface and lose the protection offered by the increased transport of oxygen in plasma that occurs at depth when the partial pressure of oxygen in inspired air is elevated.

A lower maximum carbon monoxide concentration is needed for deep and saturation divers. This is because the exposure times are longer. In addition, the oxygen partial pressure is usually limited to about 0.4 ATA, so the protection from an elevated oxygen pressure is reduced.


Oil occurring as a mist or vapour can cause compressed air to have an unpleasant odour and taste. Its direct, toxic effects in normal people are not known except that in high concentrations oil vapour can cause lipoid pneumonia. In some people, low concentrations of oil vapour can trigger asthma. Condensed oil, especially if combined with solid residues, can cause malfunctions of equipment. The other problem with oil is that it can decompose if overheating occurs and can generate hydrocarbons and toxic compounds of carbon, nitrogen and sulphur, depending on the oil composition.

Some compressed air standards distinguish oil from other hydrocarbons and specify maximum limits for each. Most hydrocarbons in high-pressure systems can be serious fire hazards. Some have other undesirable effects, such as being carcinogenic.


Control of water vapour is needed to reduce corrosion and oxidation damage to equipment. A low water concentration may also prevent ice formation and supply blockage or a free-flowing regulator when diving in cold water as a result of adiabatic cooling during pressure reduction. Some investigators think that this problem has been overstated because the areas susceptible to blockage are at lower pressure than the cylinder. In these areas, the air will not be saturated because the gas has expanded. Water condensation can also impair the efficiency of the filters used to remove other contaminants. This is more common with some of the molecular sieve filter systems.

Deaths have been reported from diving with steel cylinders containing water. Rusting occurs if these cylinders are left unused for long periods. Rusting consumes oxygen and leaves a mixture that caused death from hypoxia. The other problem is that the rusting process weakens the cylinder and may cause it to become an ‘unguided missile’ if the gas rapidly discharges. Severe injuries and deaths have been caused by exploding cylinders.

Solid particles

These particles must be controlled by filters to protect the diver and the equipment. The effect of the particles depends on their size and composition. Particles such as pollens can cause hay fever and asthma in susceptible divers. Pollens have been found inside scuba cylinders. Other particles have various undesirable physiological effects depending on their size and composition. Any dust that causes coughing could be particularly hazardous, especially for a novice diver.

In diving equipment, abrasive particles such as mineral dust would accelerate wear on the equipment by abrasive erosion. Soluble particles such as salt crystals can accelerate corrosion by promoting electrolysis. Organic dust can also contribute to a fire hazard. There have been cases of filters breaking down, letting material through and contributing particles of filter material to the air supply. Large concentrations of particulates can become a fire hazard.

Nitrogen dioxide and nitrous oxide

Some of the oxides of nitrogen, and nitrogen dioxide in particular, are intensely irritating, especially to the lungs, eyes and throat. These symptoms can occur when an individual is exposed to gas with a concentration of nitrogen dioxide greater than 10 ppm. At lower concentrations, the initial symptoms are slight and may not be noticed, or they may disappear. After a latent period of 2 to 20 hours, further signs that may be precipitated by exertion appear. Coughing, difficulty in breathing, cyanosis and haemoptysis accompany the development of pulmonary oedema. Unconsciousness usually follows.

The typical maximum level of no higher than 2 ppm is also the maximum allowed level for 24-hour exposure in some standards. If the effect is increased with pressure, then 0.2 ppm may be a more appropriate limit. In industrial cities, 2 ppm is often exceeded.

Nitrous oxide is an anaesthetic agent, but only at high concentrations. A low concentration of it is specified because, if nitrous oxide is generated within the compressor a precursor, nitric oxide must have been formed. Nitric oxide can also be converted to nitrogen dioxide at higher pressures and temperatures. Therefore, a compressor that adds nitrous oxide to the air being compressed can also form nitrogen dioxide.

Odour and taste

Odour and taste are controlled to avoid air that is unpleasant to breathe. They also provide a back-up for the other standards because if the air has an odour, it contains an impurity.

Volatile hydrocarbons

Volatile hydrocarbons such as benzene, toluene, xylene and ethane, among others, exist as gases at temperatures usual to diving situations and, as such, can be absorbed and distributed throughout the body in a similar manner to volatile anaesthetic agents. However, their side effects, such as impaired consciousness and increased cardiac irritability, present additional dangers to the diver. These gases may also be carcinogenic and present a fire hazard, so they need to be limited to a maximum of 5 ppm.

Most air purity standards do not require testing for these hydrocarbons, with the exception of the Canadian CSA 275.2 2004 standard (5 ppm of volatile non-methane hydrocarbons and 5 ppm of halogenated hydrocarbons); and the US CGA G-7.1-2011 standard (25 ppm of volatile hydrocarbons).


An experienced diver was diving in an area subject to tidal currents. He planned to dive at ‘slack water’ and anchored his boat a short time before the low tide. The hookah compressor was correctly arranged with the inlet upwind of the exhaust and the dive commenced. After an hour at 10 metres, the diver felt dizzy and lost consciousness but was fortunately pulled aboard by his attendant and revived.
Diagnosis: carbon monoxide poisoning, confirmed by blood analysis.

Explanation: As the tide turned, so did the boat. This put the compressor inlet downwind of the motor exhaust. The carbon monoxide from the exhaust was drawn into the compressor inlet and was breathed under pressure by the diver.


A 35-year-old man, with 20 years of diving experience and no relevant medical history, undertook a solo crayfish dive. He told the boat operator that he would be 15 minutes, but he failed to surface. A search by police divers found him the following day at a depth of 9 msw. The autopsy was limited because of destruction of the body by sea lice. The police investigation suggested that he was diving over-weighted with 17.5 kg on the weight belt, which was not released. All his equipment was intact and working correctly and the cylinder pressure was 194 bar, so he died very early in the dive.

Analysis of the cylinder contents revealed an extremely high carbon monoxide level, 13,600 ± 300 ppm (NZ standard <10 ppm), as well as increased levels of carbon dioxide and methane. A second cylinder owned by the diver returned similar analysis. Both cylinders were filled at the same time at the same dive shop.

The coroner’s finding was that ‘death [was] due to asphyxia due to his cylinder gas being contaminated with carbon monoxide, brought about by an idiosyncratic malfunction of the air compressing equipment’. There was no evidence of any other cylinders filled on that day reported as contaminated, so this was an isolated finding, the cause of which was unknown. (This case is from the New Zealand diving fatality data and was reported in Millar IL, Mouldey PG. Compressed breathing air: the potential for evil from within. Diving and Hyperbaric Medicine 2008;38(3):151.)